Does anyone want to comment on Feyarabend’s take on the consistency requirement for new theories?
I think he is wrong but I might be nit-picking.
Ed
Link… link… link.
If you are talking about this guy…
…I consider him to be a philosophical moron.
But then again, I find myself thinking that about a lot of people. 
He seems to have valued open-minded experimentation, even with respect to theory. And why not? It’s likely an issue of balance.
If Science doesn’t represent something specific, it doesn’t really represent anything at all.
If you are going to call it “Science”, it should be conforming to a particular methodology, just like anything else.
That isn’t to say that no one can experiment in other ways. They just shouldn’t be calling it Science.
The effort to obfuscate the meaning of the name “Science”, is an effort to destroy it entirely.
This may be true, but his argument is very interesting, actually. He demonstrates that many advances were made in science via methodologies not considered proper. He carefully documents this and further shows how these discoveries, theories and processes represented positive forward movements, not only because of the specific findings, but further because of how they influenced theory and thinking. These advances are part of the history of science and some of the scientists involved are key figures in the history of science. IOW he is not arguing that science should be open methodologically, but that it already has been in practice and that not only has this led to advances, but further the specific cases involve canonological figures within science. So the Word ‘Science’ itself has been systematically used to refer to processes that vary much more than what fits the current definition.
I’m not disagreeing (yet, anyway), but what is the current definition?
Hi to All,
James: I am reasonably certain that you would by nature adamantly disagree with Feyarabend, who describes himself, with respect to the methodology of science as an anarchist.
Anon: I am just happy that you responded to this post.
Moreno: I totally agree with your comments on Feyarabend.
Flannel Jesus: I would go about defining Science this way:
Science is a collection of scientific theories. A scientific theory is a statement, or group of statements, that make predictions about the empirical world in a given domain.
Generally:
The question then becomes are some theories better than other theories?
The following is a quote from a scientist and I believe that most scientists think this way.
“A scientific theory is a synthesis of well-tested and verified hypotheses about some aspect of he world around us. When a scientific hypothesis has been confirmed repeated by experiment, it may become known as a scientific law or scientific principle. A scientific fact may be defined as an agreement by competent observers of a series of observations of the same phenomena. From time to time scientific facts are revised by additional data about the world around us. Scientists often employ a model in order to understand a particular set of phenomena. A model is a mental image of the phenomena using terms (or images) with which we are familar. For example, in the planetary model of the atom scientists visualize the atom as a nucleus with electrons orbiting around it in a manner similar to the way that planets revolve around the Sun. While this model is useul in understanding the atom, it is an over-simplified description of a real atom and does not describe/predict all of its attributes.
A particular feature of science is that it is continually evolving as a result of the The Scientific Method which calls for a constant testing of ideas and observations of scientific facts and theories/models. In order for science to evolve previously accepted theories (PATs) must be superceded by new theories (NTs). Since it is relatively easy to make up new theories, particularly if one is unencumbered by observations, there must be criteria for the replacement of previously accepted theories by the new theories.
Criteria is the plural of criterion. a criterion is a standard of judgement. A particularly good example is the Ms. Universe (or other beauty) contest where the criteria are well spelled out and the procedures to reaching a final choice are well-defined.
Here are five criteria that are generally used when comparing theories and a new theory statisfying these will then replace a previously accepted theory.
• I. The previously accepted theory gave an acceptable explanation of something, the new theory must give the same results.
• II. New theory explains something that the PAT either got wrong or, more commonly, did not apply.
• III. Makes a prediction that is later verified.
• IV. Elegance - Aesthetic quality - simple, powerful includes universal symmetries. That is simple, easy-to-remember or apply formulation, experssed as some symmetry of nature, be powerful enough to used in many applications
• V. Provide a deeper insight or link to another branch of knowledge”
Personally, without going into detail, I think that the sentence, “When a scientific hypothesis has been confirmed repeated by experiment, it may become known as a scientific law or scientific principle”, while true from the point of view of a large portion of the Scientific community, carries no logical or historical value.
Worse is the statement, “A particular feature of science is that it is continually evolving as a result of the The Scientific Method which calls for a constant testing of ideas and observations of scientific facts and theories/models”. This is demonstrably false if one looks at the history of science (As Feyarabend did).
My particular point is that criteria I., which is generally referred to as the Consistency requirement, is valid at least in context of the quote.
If I recall correctly, Feyarabend argued that the Consistency requirement did not apply, when comparing theories, for the following reasons:
- A competing theory might give better predictions over a larger domain.
- All theories contain, what today might be called outliers or some predictions which do not match well with experimental results
- Languages do not convey the same content to all investigators.
I think that 1) is not valid because domains either are or should be specified in a theory. Just because Theory B is better in a larger domain does not mean that Theory A should not be considered preferential, all the other criterion being the same, in the smaller domain.
Feyarabend may (and I would stress the word may) be correct that all theories have some flaws, but that does not mean in aggregate, two theories can not be compared with experimental results and the best theory, at least over the tests, be determined. (In rare cases they may tie).
Languages, in general, do carry ambiguity, but in many cases the language used in scientific theory is mathematics and, analogous to the cited quote, competent practitioners of that constrained language appear to me to communicate quite well.
Thanks Ed
I agree that Science as a typical socialist structure tends to be presumptuous in its determinations. And in that regard, it certainly should be overreached. But Feyerabend is making the same mistake as his complaint. He is presuming “Science” to be whatever people respect as knowledge and thus from where ever knowledge is to come, it is to be respected as “Science”… but without specifying where it came from. The name “Science” must refer to a specific form of determination, else no one will ever be able to trust it for anything.
Science doesn’t invent ways to discover truth. Its only function is to empirically verify someone’s guess work. Science really means “try it and see”.
What Feyerabend is suggesting is that we don’t really “try it and see”, but rather, simply respect that someone, somewhere, impressed a lot of people with the notion of it being true. In effect, he is promoting old fashion religion over Science and in place of Science, in place of “let’s try it and see”.
He wasn’t complaining, per se, at least not about what I mentioned. He was noting. He noted that advances were made through methodologies and results that do not fit the supposed criteria of scientific methodology including research considered important and groundbreaking by a consensus of scientists today. Many of these advances, if not all of them - it has been a long time since I read Feyerabend - were confirmed, often much later via methods supposedly the only ones accepted.
Not as far as I can tell. I don’t think he would have said this.
Well, part of his point is also that even things that have been determined through the proper processes have also turned out not to be the case. So it is not a choice between an infallible method and a fallible one. What he wants is to consider things more tentative, but at the same time allow for a greater range of methods. He points out that this in fact led to advances in theory, practice, technology and later research in the past, and sees no reason to close the doors on all but one methodology given this.
His point is that, literally, you are incorrect about what Science has meant to humans. That they have in fact had a more varied set of methodologies and this has been effective. So it is not what Science means to you, but it is what has been included in Science. He is - or perhaps was, he may be dead - advocating that we continue to allow this flexible set to fall under the term Science.
Of being possibly true. I mean this includes things like some of Einstein’s work that was considered incredibly likely to be true, but could only be verified decades later - this had to do with gravity affecting light near the sun. That this was considered working true led to other interesting lines of research and also other ideas that could be verified.
Well, no. All the case studies he presents do not use religious methodologies nor do they rely on faith.